O comes from a profoundly amusical family. Their
home contained a windup gramophone and a violin. Both were never used.
O remembers the sharp styli for the gramophone, but
does not remember there being any records to play. The violin (belonging to
his father) was a device even more indecipherable. His father would, at
times, exhale a vaguely tuneless whistle. Not a pursed-lip whistle
(O’s mother often said that kind of whistle
was the worst thing one could do) but rather a hiss of air passed over the
tongue, brought close to the roof of the mouth, through the top incisors, and
out; an almost tuneless whistle. His mother, for her part, would occasionally
sing a single line from one popular song or another.
Perhaps the earliest sound he remembers noticing was the summer cicada.
Later there was a dilemma (similar to when his father misinformed him that
clouds were all the same height in the sky; which was contrary to what he
thought he could see)—was the cicada biological or electrical? Where
did it come from?
What other remembered sounds? A ringing bell on the milk wagon, pulled, he
remembers (perhaps inaccurately) by a snorting, clopping horse; the
calliope-pistoning of a push mower across the lawn; a phone ringing; the
metal tongs of a fan rake pulled through fallen leaves; the crackle of these
leaves as one tossed oneself into their heaps; fire in its fireplace,
thrombing, snapping wind in its many sonorities. Talking he remembers at
first not understanding; he would soon discern that his grandparents used two
languages, German and English. Later, the deep-throated wattle of the sedan
parked next door, similar to an inboard motor launch. The family lived
inland, although he learned that his parents had wooed by a small lake. And
later, one week per year was devoted to living by one of the Great Lakes,
waves; boats of all sizes. O was able to swim at the
age of three, and then he forgot how.
The neighbours next door with the throaty car also had a piano and
O received a couple of lessons there from the
neighbouring matriarch, of which he has no other memory.
The family procured a television set, and there were evenings of music
playing in the house, particularly the fiddling, dancing and singing of
Don Messer’s Jubilee. And Sundays the portable
organ warbled during church services held in the Empire Public School gym.
And everyone singing, approximately the same tune, each singer varying in
temerity and accuracy.
A new family car came with a built-in radio. This was soon followed by his
very own eight-transistor portable radio receiver. This radio had a choice of
two bands, AM and SW. O spent hours, particularly
evenings, carefully advancing the serrated, gear-like plastic tuning dial;
finding, between locations of foreign voices, many mysterious sounds to be
interpreted. The rapid pitch blipping (data transmissions) he mistook for
satellites in earth orbit; the harmonic roar (signal jamming) he interpreted
as the propeller and engines of airplanes, and he listened to long stretches
of these seldom-varying drones, waiting to hear pilots asking for landing
instructions. To pass the time, he would adjust the tuning dial fractionally,
by millimetres, and thus sweep up and down through the frequencies, and the
glissing whoops of carrier sines. In the AM band he would gradually adjust
the frequency from station to station wondering where the stubby red
diamond-shaped pointer would centre on the call numbers (1490, 1050)
pronounced by each ‘standard broadcast’ vicinity in relation to
the mysteriously spaced numbers
[ 54 6 v 7 8 9 1v2 16 (x100 kilocycles) ]
printed above the tuning trough. He’d open the back of the radio to see how the dial
and pointer were attached to a much more elegant looking turbine thing.
And then there was the volume dial. O had to open
the brown leather case to see the indication LOUD <— OFF. The LOUD
was brash; OFF was indicated by a click; and just next to OFF was a secretive
miniaturization of sound. A similar tinny tinyness could be attained by
canceling the radio’s main speaker. a paper cone hidden behind a gold
metal fine-perforated faceplate behind the coarser perforation of the brown
leather case, by plugging into the 2nd of two receptacles provided for the
earplug speaker jack attached to a white-plastic-coated pair of wires
attached to a white pill attached to a white finger with a nostril at its tip
— the tiny sound would emit from the nostril; or, if the plastic finger
was fit into one’s ear, the sound was louder, and private. There was
only one earspeaker provided with the radio so O
would have to choose which ear to plug. Left or right? What was the
difference? Why, when the voices from the radio were as loud as himself
talking it seemed way too loud, even though his own voice was not a very loud
one.? He mumbled. He still has this radio. It still works, although the once
teaming short-wave band is now almost deserted. He still mumbles.
The next device of disembodied sound to enter his life was a four-speed
automated phonographic record player in a cardboard case. the four speeds
were 16, 33 1/3, 45 and 78. The phonographic records purchased for
O were labeled 33 1/3 rpm and 45 rpm, but there was
nothing to limit any of them from being played at any of the four speeds. The
transition from 33 1/3 rpm to 45 rpm transposed the human voice to chipmunk
range, an effect popularly exploited by rodent puppets, as he heard on the
Ed Sullivan television show. The more evocative
transformation of a woman’s voice into a man’s was simulated by a
transition from 45 to 33 1/3rpm. This and the approximate doubling in speed
and duration to 16 and 78 eventually led to some of O’s favourite plunderphonic effects, and the concept of
harmonic speeds.
This 4-speed time machine would eventually be replaced by a hi-fi
two-speed semi-automatic phonograph, which was followed by a two-speed manual
turntable. These successors eschewed the exotic speeds of 16 and 78.
The device which came next into his life was a solid-bodied guitar. It was
designed to be magnetically and electronically amplified but O never obtained the requisite amplifier, so this guitar’s
dynamic range was restricted to (but not limited by) the approximate range of
a clavichord. Each week a guitar teacher would plug this guitar into an amp
and tune it. After a few weeks, the guitar teacher dissuaded O from further supervised study. O
never tuned the guitar again but continued to play it for years, unamplified.
It was quiet. It was private. It still has some of its original strings.
The unamplified guitar and the record player were the instruments which he
brought to his first musical ensemble, a mixed-media quartet. The other
musicians were boys from the same block as O. they
rehearsed in houses on the street and performed in the backyard of
O’s house. The other instruments were trumpet
and bongos.
O is not sure which noise-producing device came
next, a Spanish guitar with nylon strings, or his first tape recorder. He
remembers working the two together: dropping the tape recorder microphone
between the strings, into the hole of the guitar; then shaking this
concoction and recording it; and playing that back, backwards.
The tape recorder was a portable, monophonic reel-to-reel, a present from
O’s father, who repaired typewriters.
O soon modified its recording function by disabling
the erase head, a magnet on a movable arm. With the eraser out of the way he
was able to record layers of time. To the sound of the shaken guitar he added
the sound of the strings being scratched by various things. The heads of the
tape recorder were half track: designed to play the tape through in one
direction and then the ‘flip side.’ But O wanted to go backwards through time. Until he discovered how
to wrap the tape deviously around the motor capstan, he would twist the tape
around moebiusly and listen as the reversed sounds muffled through the
acetate backing to the magnetic layer. Sounds would present themselves tail
first, but it was particularly with reverse talking that he found a foreign
language of interest to him.
O began to create fictionalized musical
documentaries. He once taped himself and a friend playing penny whistles and
household percussion, and presented the result in music class at school;
passing it off as a field recording from the jungles of Africa lent to him by
an uncle on safari.
O began planning technically idealized pop songs,
which would, for instance, have tempi in simple rations to 33 1/3 and 45 rpm
— the result being that vinyl scratches would be rhythmically
co-ordinated and would not phase in relation to the rhythm of the
content.
Eventually O entered high school, where he was
not accepted into (nor allowed to audition for) the school band. Using an
insider as a conduit, instruments were borrowed or purloined, and
O started his second musical ensemble. At first this
fringe school band was open to all comers. Their intention was to play rock
music (with tubas and timpani!) but the ensemble’s arrangements were
never other than entirely improvised. Since then, O
personally has always approached the performance of music and dance as an
improviser.
And that’s another story.
Meanwhile, after leaving high school, O briefly
attended several universities. He traveled for the first time to the West
Coast to study Communications with the composer R Murray
Schafer, and consequently worked with Schafer’s World Soundscape Project for a summer. While
there, he created several projects in the WSP electronic sound studio,
including the first of many swarmfields (sounds time-staggered and
multi-tracked so many times that they become drones) and sculptural sounds
(drones that vary relative to the aural equivalent of viewpoint —
hearpoint?). He also developed his Burrows series, a
group of short compositions using the reading voice of author William S Burroughs as the sole source material. With the
exception of a film he made in high school, now lost, these would be his
first examples of the massive and meticulous editing and reorganization of
time. In one, for example, a mediation between backwards speech and forwards
speech was presented. Another featured an acoustic palindrome. The pieces
were all intentionally reversible and O briefly
published them on full track reel-to-reel tape, so the listener could explore
the reversions.
These compositions were part of O’s early
explorations in plunderphonic procedures (documented in the 69plunderphonicS96 box set — a retrospective of compositions
from 1969 to 1996), which relied exclusively on transformative quotation.
O has seldom been one to disguise his sources.
A little later O began developing, with
Marvin Green, Pitch,
presentations in absolute darkness. Their perceptual research manifested
itself in performances, sometimes with, and sometimes without aural content,
curated variety concerts, with ushers and performers using elaborate systems
to ‘see,’ while the audience remained in utter black.
O also created, with Emile
Morin, a Pitch installation, for one participant at a time, which
encouraged an absolutely sightless journey from hard gradually to soft, and
from hearing the outside world to an entirely insular experience.
From then on the ongoing obsessions with improvisation, perception, and
transformative appropriation would continue as independent, parallel
investigations.
During the last quarter of the 20th century O
founded and developed Mystery Laboratory (a.k.a. mLab) a perceptual research
facility and recording studio in a small attic atelier, in Toronto. At first
there were no multitrack tape recorders in the studio. O preferred the way film sound was put together on sprocketted
reels, each of which could be time-shifted in relation to the others; but
instead of sprocket-synch, he preferred wild synch, in which various tracks
of sound, on independent machines, would be started in relative
co-ordination. Synchronization was approximated in performance.
O continued to exaggerate the time-base of
existing music; either slowing pieces down a dozen times, or speeding them up
to the spectral threshold of audibility, particularly with mono-harmonic
examples such as a raga or the opening of Mahler’s First Symphony.
With the acquisition of some synchronized multitrack recording machines in
the mid-’80’s and computer audio technology a few years later
O entered a renaissance period of plunderphonics
development. Many of the techniques he developed were applied to a broad
range of popular music and the techniques of popular music were applied to
less-than-popular fare.
Norman Igma [vi-03]
Interview with John Oswald (part 1 of 2)
by Norman Igma|imed_0368-0002|
In the autumn of 2002 mLab observer Norman Igma
spoke with John Oswald about aparanthesi
i: We’ve talked in the past about your activities
in terms of categories, such as improvisation, plunderphonics, discosphere,
etc. This will be the first time you’ve asked me to discuss with you a
single composition. If one were to take at face value your assertion that
aparanthesi consists of just one note, then it would
seem that there isn’t much to talk about.
o: I thought you might be interested in discussing
something simple in great detail…
i: … with the result, I expect, that it will seem
less simple. So, where would you like to begin?
o: Well, let’s begin at the beginning. But not, for
now, at the beginning of what’s on the CD, particularly since there are
two beginnings, and I would first like to describe my concepts at inception,
which I think will help explain why there are two versions of more-or-less
the same thing.
i: You received a commission from Réseaux [with support
from the Canada Council for the Arts] for its “Rien à voir”
concert series in Montréal…
o: … to create an electroacousmatic composition of
approximately fifteen minutes in duration for their festival in the winter of
2000.
I usually already have an idea of the tangent I would like to take when I
accept an invitation, and in this case it was the one-note idea. I was hoping
to find an interesting way to follow a single specific pitch gradually
through many states; it would be, in a sense, the morphing of a pitch through
various instrumental, acoustic and electronic forms.
I had often before explored slow transformations, particularly in
2 Crossfading Stills and a Shifting Multiple
Transparency [1986], and the sonic image of a concert pitch 440 Hertz
sine tone transforming into an orchestra tuning has been used to begin
several of my pieces, including WX [1980],
’Plause [1986] and Tune
[1989]. Minus the sine tone it also is the beginning of my first string
quartet, Spectre [1990] and my first full-orchestra
piece, Orchestral Tuning Arrangement [1994],
co-composed with Linda Catlin Smith.
It’s almost the same gesture in each of these pieces: a single
uninflected A, whether it’s begun with a sine tone or an oboe or a
cello, transforms into a broader timbral spectrum, with the A as the spine.
That thickening A is all that happens in Spectre,
which was composed for the Kronos Quartet with a
thousand recorded images or reflections of a string quartet playing an A
which is very gradually inflected with an increasing degree of subtle tremolo
and vibrato, which then slows and broadens into wandering glissandi.
Spectre has the same macro-wave form as my analogue
tape meditation on fire Skindling Shadés [1987], which
also features this wandering-gliss effect multiplied hundreds of times, to
create a rather massive sine composite. I first began this approach, which I
called swarmfields, around 1973 or ’74.
A pure sinusoidal tone is my favorite electronic sound. In fact it’s
just about the only electronically generated sound that I ever use in making
music.
i: So you’ve been attracted to 440 Hertz pure tones
for quite some time.
o: Another oft heard tone is 1 kilohertz. This is a
common test signal. That’s the signal broadcasting stations play when
they’re off-air. It’s a soprano pitch. 440 is alto. It’s
the pitch reference that almost all music, of the sort that has been
influenced by equal temperament, use as a reference. But it’s not what
any of the worldwide electrical grids are tuned to. That would be 480 Hz
(which is 3 octaves up from the 60 cycle hum we often hear in North America)
or 400 Hz (in Europe) — about a quarter tone sharp and flat of 440, or
A. So quite often music is out of tune with the noise generated by amplified
or electronic musical instruments. It’s out of tune with the lights and
ventilation in concert halls.
We can come back to this when we talk about version B of aparanthesi, which is tuned to the North American electrical
standard. In any case I initially followed the concert norm and tuned this
piece to 440. I do try to be conventional when there’s no good reason
not to be.
Now when I say that aparanthesi is one note I mean
that it is one specific pitch and manifestations of that pitch in the various
audible octaves. For A there are ten audible octaves, from the lowest note on
most pianos (A=27.5 Hz) to 5 octaves above A 440, which is 14,080 Hz. These
ten manifestations of simple halvings and doublings of a set frequency are
the entire harmonic content of aparanthesi — any
deviations from this frequency gravitate back to it, by various means.
i: Tuning.
o: Yes, tuning. There’s are two human solo
instrumentalists in this piece. One of them is an anonymous piano tuner. The
other is a cellist, Joan Jeanrenaud. And then almost
everything that has a broad range of frequencies, or noise elements, is
gradually filter-tuned to this tonic. A thunderstorm becomes a thunderstorm
in A. All the birds in an aviary sing in A. The 88 notes of a piano are all
tuned in forced perspective to one A.
i: How did this odd instrumentation of piano and cello
and birds and weather…
o: … and a foghorn and cattle.
i: Cattle?
o: Yeah. There were some distant cattle in the source
recording for the thunderstorm. It was a prairie storm. I tuned the moos to A
and then I changed my mind and suppressed the cattle sounds; they were
distracting. And I forgot about them until just the other day I heard them
again when I was listening closely — I have a pair of in-ear monitors
which have been custom-made to sit about half a centimetre from my eardrums.
It’s hard to miss anything when listening with those and the cattle are
still there, in both versions.
Which leads to the subject of subliminal images and sounds that are at the
threshold of audibility. They are not necessarily quiet because they can be
buried in a loud sound. When I was first formulating this piece I was most
interested in creating audio morphs; an identifiable sound which would
gradually transform into something intangible on its way to becoming a
different identifiable sound. I got particularly interested in that
intangible area, by which I mean when a sound on its way to changing into
another sound, becomes something in the middle which is neither the original
sound nor the destination sound, although the listener might already guess
that destination sound, or they might begin to hear something similar which
is neither of those two sounds.
I had been researching the various attempts that were being made to
metamorphose sounds — to, in effect, create an acoustic analogy to
morphing robots and the effects that happen to women’s coiffures in
hair conditioner commercials on television. I wrote to a few
pyscho-acousticians and programmers who have worked on this, but no one
seemed to be very interested in my situation. Eventually I figured out that
this was because, in cases, such as mine, where there is no pitch change,
morphing can be accomplished by simply adjusting the dynamics of the
harmonics. For instance, by applying the shape of harmonic spectrum of a dull
sound as an envelope filtering a rich sound, you can make the rich sound
imitate the dull sound. Morphing two sounds of the same pitch is simply a
matter of crossfading over time from one to the other. If this is done with
active filters (I mean filters which progressively emphasize or suppress the
frequencies which constitute the characteristics of that sound) one can go
from a sound to another in a really lovely gradual sort of way.
The original premise of the piece was that there would be an initial
sound, probably a sine tone, same as I’ve used in the past, and this
would transform into the wind which would transform into a flute which would
transform into a highway heard through a narrow drainpipe which would
transform into a human voice, and so on through a multitude of
transformations.
I started by making a catalogue of orchestral instruments playing steady,
sustained A’s, sans vibrato, and we (Phil
Strong did quite a bit of work on this) expanded this catalogue by
adding these instruments with their different natural pitches shifted in time
so that they also became A. Then I did some of the morphing technique I just
mentioned. I particularly enjoyed working with winds, because, in listening,
I could imagine the physicality of the instrument transforming, so a conical
clarinet would curl into a French horn for instance, and then straighten out
into a didgeridoo.
This sort of thing occurs as a quick transforming figure in aparanthesi [A.12:20->12:50 or B.30:00->31:10], and very
slowly in the voice horn fog [A.13:24->15:40 and B.15:48->19:45], which
I’ll talk about when we go through the tracks sequentially.
i: The orchestral instruments and voices all seem to
blend together, but the piano sticks out, which makes the whole thing feel a
bit like an electroacousmatic piano concerto.
o: In 1999, when I was finishing up aparanthesi I was also beginning to work on what I call The Idea
of This, for the Glenn Gould ballet, Disembodied Voice [1999], choreographed by James
Kudelka for the National Ballet of Canada. The starting point for that
was my plunderphonic creation Aria [1989], based on Glenn
Gould’s final recording of the intro to the Goldberg Variations. It was interesting to hear, in Aria, how a particular machine would interpret Gould’s performance somewhat inaccurately, and I liked the
resulting music a lot. Then subsequently, a composer acquaintance of mine,
Christos Hatzis, employed a fellow named
Ernest Cholakis to attempt to make an accurate
MIDI transcription of the Goldbergs. I
was curious to see how far one could go with this. I had the idea that there
could be a, in-a-sense restored recording of this performance which would not
be heard through loudspeakers, but through a piano which would be played by a
reproducing mechanism, and ideally that that piano would be the same one that
Gould made the original recording on.
Quite conveniently Gould’s piano was a
Yamaha, and Yamaha has developed a line of player pianos called Disklaviers.
Gould’s Yamaha is on permanent display in
Toronto, which is where I live and work. I set about to try to convince
Yamaha, the Gould Estate, and a bunch of other
people to convert the Gould piano into a Disklavier,
so that it could be more of an active musical museum piece, rather than an
inert one. This was a long shot, and in the end it didn’t happen, but
meanwhile Yamaha graciously lent me a decent Disklavier grand for a year, and
another one for Ernest, who perfected our reproduction on a piano at least
similar to Gould’s.
So all the piano sounds in aparanthesi come from
this Disklavier, which arrived towards the end of the composing of it.
i: Did you know before then that piano sounds would be a
pervasive part of the composition?
o: I don’t think so, although I might have. I seem
to remember that I wanted to start the piece the same way as the
aforementioned compositions WX, ’Plause, etc., with a sine tone fading in and becoming
gradually harmonically richer. As it turns out a big piano chord, [A.00:00
and B.00:00] is my other oft-used opening strategy; a piano chord (an
electroquotation of the end of The Beatles’
A Day in the Life) is the beginning of the
Plunderphonic CD [1989] as well as its successor, the
69Plunderphonics96 box set [2000], and I reversed it
for the beginning of one of the dance scores I made for choreographer
Bill Coleman [1993]. Also the big strummed piano
chord at the beginning of the Gordon Monahan CD I
produced [1992] is one of my favorite beginnings. That disc in particular
increased my interest in the acoustic image of the piano. A piano is
difficult to record because it is big. The classical solution is to always
record a piano in a room. This also applies to recordings of orchestras. The
listener is never really close. We were able to record very close for the
Monahan disc particularly because of the material,
which focused on particular regions of the piano, sometimes a single pair of
strings. I developed a physically and electronically moving soundfield
microphone technique which we tagged O.M.N.I.V.E.R.S.E. [orbital microphonic
navigational imaging via ecotonic radial stereo eccentricity]. In effect we
treated a piano as if it was a landscape in a film. We panned, tracked and
zoomed continually over this pianoscape. The effect is aurally much more
subtle than you would imagine. Our achievement was to record a whole piano
the way pop crooners are recorded — in-your-ear close-up.
This sense of landscape, or more precisely soundscape, was present
throughout the composition of aparanthesi. This
particular soundscape is both inside and outside. Many of the sounds occur
obviously outdoors, such as thunder and birds. Some of them, for example the
rain storm, were recorded from a sheltered position, and then the piano,
other instruments and voices were recorded in small isolated rooms. Because
most often these inner and outer states are occurring simultaneously, and
particularly because all the inside sounds were close-miked, which
de-emphasizes room ambiance, there is little or no apparent division between
environments. I was consciously trying to make it seem as if the
listener’s perspective is not moving around, but, instead, different
states are evolving from one point of aural perspective. This is somewhat
akin to what Stockhausen described as seeing the same thing continually under
differing light, rather than the usual cinematic method of seeing a variety
of things under the same light.
There’s another angle that comes to mind about this outside/inside
thing, which is related to the first public audition of aparanthesi, which took place at the Ex-Centris cinema in Montréal
for the “Rien à voir” festival.
i: Which is the same as track 2: aparanthesi B on the CD.
o: Approximately yes. aparanthesi
A [track 1] is the final revision of this piece, idealized for home
listening. It is the result of almost two years of sporadic revision of the
composition.
aparanthesi B [track 2] is a dynamically compressed
stereo realization of the eight-channel concert version of this piece which
was created with that particular premiere performance in Montréal in 2000 in
mind. It had a different title back then.
i: What was that?
o: It’s gone through more titles than anything else
I’ve created. The most pervasive one during its conception has been the
symbol A). The title aparanthesi is a slightly
modified spelling out of that equation. But for some reason that escapes me
now, for that performance only, it was called Just or Jeus, depending on who
you ask.
But before that, the festival organizer, composer Robert
Normandeau, seemed to be slightly amused that I made a special trip
from Toronto to Montréal two weeks before the performance, particularly to
observe the characteristics of the lobby and the entrance way to the
theatre.
This is what we did for the concert. The first part of the programme was
an appropriately psychedelic diffusion by Julien Roy
of the Cease Tone Beam section of Mirror Ashes from my Grateful Dead set
Grayfolded [1995]. The audience was then invited to
leave the theatre. In the lobby, which is a rather large multi-storey atrium,
I had the actress Pascale Landry standing on a
loudspeaker pedestal facing away from the audience. She was very, very slowly
turning towards the audience. There was a very slow crescendo of a drone
which turned out to be a reversed reverberation, or preverberation, of her
speaking a single syllable, “la.” This was followed by
progressively shorter preverb ramps to progressively closer spaced syllables
and words of a text which she appeared to speak eventually in a continuous
stream without breathing, by which time the reverse reverberation had
disappeared. The preverberation resonated from loudspeakers in the upper
balconies of the atrium. Pascale’s voice came from the speaker she was
standing on. In fact what the audience heard was a recording of her voice to
which she was very precisely lipsynching. If she hadn’t been
pre-recorded, it would have been impossible to anticipate her vocalizations
for the preverb and it would have been impossible for her to speak
continuously without breathing.
i: You cut out the breaths.
o: In fact each word of the text had been recorded
separately, so all those spaces had to be cut out.
Anyway, following this performance the audience returned to the
auditorium, but the doors were closed. Ushers then opened the doors precisely
synchronized with the sounding of the initial chord of the piece, which
erupted from the doorways. In the theatre itself this sound was at an
extremely high volume. It would have been unpleasant and perhaps harmful to
experience this degree of volume directly, but from the lobby one got the
impression that something powerful had just occurred in the room.
We replicated this beginning for a hearing a year later [in 2001] at the
Festival Archipel in Geneva.
The chord itself consists of the eight A’s on a piano, including its
lowest note, played precisely synchronized on the Disklavier. And for this
performance everything was pitched up to approximately a quartertone flat of
a B natural, which was my choice of tuning for that room.
Then the audience entered the aftermath of the initial chord and found
their seats. In the B version the piano decay is fairly quickly replaced with
a ten-octave sine-tone chord. Each of the tones independently and slowly
varies in volume, sort of a one-key sine organ, with variable stops. Because
the tones themselves are pure and absolutely in tune with each other, the
chord sounds remarkably still, but the audience is moving, because
they’re still getting to their seats. It’s a period of
adjustment.
One thing that did happen in an obvious way in the concert version was
that things became gradually very quiet. I worked on the first third of
aparanthesi for several months. The piece was
auditioned with an extreme dynamic range. And most of that extreme was
extremely quiet. Most CD releases use very little of the 90 dB dynamic range
available in 16 bits, except when there’s a final reverb decay. I was
working with the increased dynamic range of 24 bits, and I was very
interested in having things happen at the threshold of hearing, where and
when you are not certain if you’re hearing something or imagining it.
So I was working with a very precisely set range of dynamics in a fairly
quiet environment, and I would bring people in to test what they were
hearing. The concert dynamic was also quite extreme. The initial sound was
huge, and then by the time the audience was settled in their seats they may
have not been sure whether the piece was still playing or not. There was also
the low level prominence of the highest tone which would be 15,360 Hz, which
some of the older members of the audience, myself included, may not have been
able to hear but which, by the very fact that it was moving air, could be
sensed in some way.
For this CD release I have no way of instituting set playback levels for
every listening situation, nor would I want to, so I’ve reduced the
dynamic range considerably, to allow playback situations to be more flexible.
I’ve also reduced high and low pitch levels relative to the general
increase in volume in proportion to Fletcher-Munson hearing sensitivity
curves.
Norman Igma [vi-03]
Interview with John Oswald (part 2 of 2)
by Norman Igma|imed_0368-0003|
i: Didn’t you say that the piece also divided into
eight channels then?
o: Yes, and ideally I should have used more channels than
that, so I could have assigned each of the sine tones to its own speaker and
spatial path. I’m not going to go into the multi-channel aspect much in
this conversation, partly because it was a subsidiary aspect of the work
which I didn’t follow through on to the extent that I perhaps should
have, so when I went back a year later to try to spatialize the piece I found
it frustrating. I think the closest to ideal versions of this composition are
these stereo realizations. I will come back to the difference between
multi-channel and stereo in a moment, when I describe the coneflapper
[A.02:30->04:00 and B.5:40->6:45]. First let’s finish talking
about the opening gesture.
In aparanthesi A the initial piano chord is
reinforced both by the thundercrack and by an extension of the piano sound
through conventional reverberation and by more extreme sound freezing
schemes. The piano itself is naturally not as in tune as the sine tones and
there are non-linearities built into conventional reverberation algorithms to
make them sound more natural. This more unsettled quality I think goes nicely
with the rumbling of the thunder decay, which almost segues into the
crescendo of the coneflapper.
The coneflapper was created by overdriving the pedal tone [27.5 Hz]
through a subwoofer; physically massaging the distortions with my thumb on
the cone of the speaker, and rerecording the result. Initially, in B, the
coneflapper was a monophonic signal which was directed exclusively at a high
level to a special small speaker I placed in the centre of the concert hall.
The intended effect was that that speaker was being seriously overdriven and
was coming apart. When I reworked this sound for A, I decided to make the
instability a stereo effect — like the giant hummingbird [A.18:09], it
won’t stay still.
The coneflapper apparently rides off into the distance — I almost
didn’t say that because in fact it’s just a dynamically filtered
fade with no echo-location cues, but in the context of the scene I’m
trying to set, which is of a very open space, this would be consistent with a
sense of retreating into the distance. I suppose the stereoization of the
coneflapper in fact reduces this sense of breadth — in free space
everything is mono.
The inspiration for the aftermath of the coneflapper was seeing the
Hollywood disaster farce Earthquake in an old theatre
in 1974. The film was billed as being presented in Sensurround. Sensurround
was the addition of a special subwoofer in selected cinemas. This was a two
or three-metre square wooden baffle in the back of the theatre. This
subwoofer wasn’t connected to the movie soundtrack; instead there was a
dedicated noise generator which the projectionist would flip on during the
earthquake segments of the movie. And the major earthquake segment lasted
continuously for almost ten minutes. The effect was quite intense, you shook
in your seat. I was surprised that we weren’t rained upon by bits of
the ceiling. I worried that the two elderly ladies sitting near me
weren’t going to make it through alive. And then it stops and you hear
some cellos playing a quiet pedaltone, embellished by a couple of harp
figures. Of course this was mixed in with the physiological results of ten
minutes of acoustic bombardment so there was a very peaceful ringing
quietness.
This is the sort of setting that I’m trying to evoke in the wake of
the coneflapper. The sine organ is still there [A.03:40->06:50 and
B.06:40->09:34], in completely different proportion, with emphasis on the
high A [14,080 Hz] and a touch of a comb-filtered allusion to one of the
swarmfields we’ll encounter in the final section.
If you’re following the timings of these two versions you’ll
see that I’ve time-compressed this whole first ten minute selection by
about a third. The B version would be even longer if it was pitched down to
A. (Listeners with pitch tuning controls on their CD players can, of course,
try this.) When I first formulated the first section I wanted it to be, at
least on a casual listening, extremely long and apparently tedious, with the
idea of picking up the pace gradually through subsequent sections of the
composition. But in the end the majority of events take a long time
throughout. So when I had this perspective I went back and proportioned the
first section [A.00:00->06:50] so that each subsequent section should feel
a bit longer, as if maybe time is slowing down.
The aftermath of the coneflapper subtly settles down [from
A.03:40->04:15] and then there’s a long slow crescendo and
accumulation of elements [A.04:15->06:50] which is inspired by orchestral
bits like the overture to Wagner’s
Lohengrin or the Buffalo
Springfield’s Expecting to Fly. There are
various voicings which are progressively added at the threshold of audibility
which resonate sympathetically at increasingly more obvious punctuation
points. The first [at A.04:33] is just a slight hump of the 440 Hz sine tone
and then each subsequent one [880 Hz at A.04:43 and 220 Hz at A.05:04] sounds
more articulated and [as of A.05:30] obviously piano-like. The notes over the
next bit more or less take turns being exciters or dampers for the sine
tones, until the final arpeggio of piano dampers [at A.06:50]. In the B
version [B.09:34] this is a sounding arpeggiated piano chord in reverse which
is meant to foreshadow the crescendo to the final chord [B.31:11].
By different means in each version we are brought to the first moment when
there’s a rest from tone, and here we are introduced to the piano
tuner.
We went through several tuners for the Disklavier, and unfortunately I
can’t remember who this one is. The recording was made surreptitiously.
In B you hear a straight documenting of some tuning preparations
[B.09:34->10:39] along with all sorts of aural room cues, particularly [at
B.10:24->10:39] where he shuffles away from the piano to unzip his case to
get some tools (there’s a resonant high/low piano interval [at B.10:22]
— it sounds like the high note was superimposed later, which
unfortunately, I now tend to think, distracts from the documentary effect).
Given that the whole first section occurs in outdoor, or free, space, I at
the time wanted this interlude to sound as much as possible like a contrast,
casual, perhaps even a different place. It’s not an absolute arbitrary
jumpcut [at B.09:34] though, because the piano elements both before and after
link the two scenes, but I think that, for a moment, before he steps on the
pedal and you hear the piano again, that it sounds like something has ended
and you’re hearing a bit of unintended shuffling around, as if you were
between movements in a concert.
And in fact another guiding element in this early version of aparanthesi was concert ritual, particularly the orchestral
tune-up, which you hear later [B.29:56->31:11].
i: I really like the point in this version where the
tuner pumps the sustain pedal of the piano [B.09:53] and you hear the first
distant rumblings of the coming thunderstorm…
o: … as if the piano activity is initiating a
meteorological response, or that there’s a synchronicity happening. And
then by the time you get to a rather obvious little edit [at B.10:39] before
going into the hammerdown, the gradual resonant focusing of the storm by comb
filters is just becoming apparent.
In the new version of aparanthesi this interlude is
handled quite differently. The roomsound clues have all been eliminated.
After what is almost like a sigh by the tuner the pedal pumping soon begins,
accompanied by a variety of piano string articulations.
i: And a frightened bird taking wing [A.07:30]…
o: Which leads me to another digressive anecdote. This is
one of the last things I added to the piece. I wanted to put some avian
indication earlier in the piece than the arrival of the tuned flock [A.15:35]
and the giant hummingbird [A.18:09]. I had a vague image of an albatross,
very large slowly flapping wings. Sound designer Peter
Thillaye, who had supplied me with the hummingbird source, suggested
we visit a foley artist named Andy Malcolm, who had
converted several buildings on a farm outside the city for the expressed
purpose of recording sound effects for movies.
Unfortunately Andy’s collection of bird wings had been recently
decimated by the arrival of some moths hidden in a hawk specimen. We worked
on the big wing sound using gull wings I think (I had to conduct because he
wasn’t used to working without a moving picture) and we got some okay
results but nothing immediately satisfying. As a way of, I guess, letting off
some steam Andy began to make these little haiku of bird activity —
brief take offs and landings. I remarked that that sound reminded me exactly
of a scene in a movie I had heard fifteen or so years years ago, Agnes of God [1985], in which a bunch of doves flocked about in a
belfry. Peter Thillaye laughed at my recollection
and said that he and Andy had done the sound for that film.
Anyway I thought the piano tuner could conceivably have something up his
sleeve, and presto…
The hammerdown, which is again from the documentation of the piano tuner
adjusting something in the piano is inspired by another recording image. In
1974 the World Soundscape Project created a series of radio works for the
Canadian Broadcasting Corporation (CBC), our
national radio. Barry Truax did one with the theme
of time. The part I particularly remember is a segue from descending
footsteps on a wooden stairway (it’s interesting to determine what are
the cues which enable us to hear the difference between descending and
ascending footsteps) to a slowing of the recording of these footsteps as they
continue to descend to an emptying drain slowing, to the tocks of a
grandfather clock which is also slowing down and then it chimes very slow and
getting lower and slower [this sequence recurs on the mystery tape GX].
The piano tuner hammering on the soundboard is mostly non-pitch-specific,
although there is a bit [A.07:51>08:01, B.10:59>11:08] where the
strings sound in descending pitches in B, which I think I’ve fixed in
A; there’s a sense of getting bigger and slower and more thunderous.
Each time the hammering hits a nodal A there’s a reinforcing note from
the next lower octave, until we hit bottom.
The idea of a harmonic relationship of different speeds of playback of a
recording is realized very clearly in another piece of mine, Bell Speeds [1983, 1990; available on empreintes DIGITALes, IMED
9604] which is also all in octaves. The chromatic version of this technique
is coming up in the third region of aparanthesi
[A.16:06->25:01, B.20:35->28:12].
Meanwhile, during the hammerdown, the storm has begun to gradually
resonate in tune. I programmed some comb filters with which I could
individually vary the prominence of each of the teeth of the comb, so I was
able to play the storm in a similar way to the sine-organ we heard earlier.
The resonant storm sounded pretty interesting on its own, and the event
envelope of that particular storm has determined the pacing of this section.
And then I’ve added a few flavours. It’s raining on the
piano.
i: The rain drops begin to sound like piano notes.
o: Ah. Well this goes back to the initial klang (to use
Tenney’s term) of the piece. As in the
homogeneity of pitch choice, I think any homogeneity of sounding sources is
good.
There are two harmonic glisses in aparanthesi A at
this point [A.10:30->11:15, 11:22->:52] which are like curtains of
timbre dropping over the storm resonance. Once I got involved in the image of
this landscape I began trying to paint in aspects with sound, and this aspect
is like those sheets of rain one occasionally sees traversing the distance.
There’s then a distinct dum-dee-dum rhythmic figure [B.15:07] which was
naturally occurring in the source storm. In B I use it to initiate a drone,
but it sounds so artificially played, even though it’s purely filtered
thunder, that you can hear the extent to which I decide to de-emphasize it in
A [A.12:22]. And then there’s a third harmonic gliss
[A.12:49->13:22] followed by [at A.12:38] a string tremelo which will be
repeated later [at A.19:30]. Then there’s feminine vocal formants
followed by masculine followed by a reiteration of the initial piano chord
which initiates the man/fog morph [A.13:24, B.15:58]. The piano chord
doesn’t happen at that point in B but I thought it was a good idea to
have that pianistic reminder at this point, to not stray to far from any of
the principal elements; and, also, I thought that male “Ho!” in B
was too much of a new thing which doesn’t get emphasized or reinforced
elsewhere in the piece. In fact there are human voices throughout the piece,
subliminally. Anyway, in B [leading up to B.15:58] there’s a clearer
exposition of that single wind instrument morphing line and then the morph of
the male chorus into the foghorn [A.13:24->15:35, B.15:58->19:40] is a
much slowed down morph. This was originally considerably longer than even the
four-minute version in A. I wanted to take as much time as necessary to give
the listener the impression that nothing ever seems to be changing, but then
you come out of it in a radically different place from when you went in. But
this seemed empirically like too long a time for the rest of the composition;
I’ll do it again someday with different elements as a separate piece
and I’ll make it real long. For now hopefully I get away with rushing
through this transition in two minutes. The disappearance of the rain is a
rather obvious fade, but perhaps not that noticeable because it’s not
the focal element. In any case, when you come out of this drone the storm is
over, the rain has stopped, and perhaps the sun is out, because, for the
first time, birds are singing.
i: This is the particular moment in aparanthesi that does sound like a scene change. Like, during the
man/fog morph we went somewhere else. The tonality even changes with the last
note of the foghorn.
o: That doppler reflection. In aparanthesi B you are once again alone with the piano tuner. He
sighs again, perhaps because he has to retune that note to concert pitch yet
again. I added the surf and leaves, wet and dry to aparanthesi A. I like how the frequency sweeps down with the
downturn of the foghorn. In either case our tuner returns us to our
tonality.
The birds were initially tuned by Phil Strong.
It’s a fairly rare bird that only sings one constant pitch, so the
choices here are all relative — they are relatively tuned to the tonic.
I don’t know how much of Phil’s set I ended up using. It’s
a fairly subtle effect, but I don’t think you’ll find a more
harmonious bunch of birds anywhere.
i: Do you mean each bird was tuned individually?
o: Yes there are a few other examples of this kind of
swarm field assemblage elsewhere. We’ll find some of this multiplying
of individuals into a mass coming up in this section.
Another transition is from day to night over the next couple of minutes.
Night creatures are introduced and the number of birds increases until it has
more of a nocturnal ambiance.
Now in B there’s a silent piano tremolo which leads up to the
forced-perspective 88-keys-of-one-note piano (that’s the thing which in
the beginning sounds like a three-inch long guitar). The silent piano is
another function of the Disklavier. By flipping a switch you can introduce a
felt rail between the hammers and the strings which prevents the two from
meeting. You get a tuneless thump. At the same time there is a MIDI piano module which is engaged which allows one to hear
the notes played via, for instance, headphones. But the silent hammering on
its own can be quite effective. This is a one note tremolo, where you can
hear the string being engaged near the end.
Also coming in from the distance is a mosquito. This is a stock sound
which comes from a commercial sound effects collection. I’m of the
opinion that it’s a real mosquito. Others insist that it is an
electrical device.
The mosquito fluctuates around the tonic, until it seriously begins to
lose fly power, by which time we’ve been introduced to the forced
perspective piano. In B it’s just the notes below the initial one, but
in A you first hear the lowest notes of the piano. Each of the 88 notes will
be pitched up (if it’s below) or pitched down (if it’s above)
that one A. It’s the kind of scaling that occurred in Bell Speeds where a tiny bell the size of a pea could be made to
sound huge. Well here you have to imagine an exact 1/8th scale replica of a
grand piano — it would be just over ten inches long. The initial sound
you hear (at A.16:06) is the lowest A of that virtual miniature piano. Then
as we work our way note by note up the keyboard, each time the piano gets a
little bit bigger, so each string stays at that constant pitch of A. As you
go above that central A the piano gets bigger, until it’s over one
hundred feet long for playing the high notes.
Now if this was a straight forward exposition of what I just described
there would be 88 notes. In fact it is once more a documentary of a tuning
procedure and some notes are played more than once and so there are more than
88 notes. And there is muting of individual strings so often you’re
only hearing one string of each set of three which represent each of the
lower notes. Plus most of the strings are actively being tuned so you can
hear the fluctuations around A, like the mosquito.
And then there’s Joan Jeanrenaud playing the cello. It’s
already been there for awhile when you hear her take a big dip [at A.16:50]
on the cusp of a high string tremolo. In fact it’s a retrograde of one
of the takes we did for Spectre [1990]. Joan’s
instruction was to play an A uninflected, and then to very very gradually
introduce a little bit of vibrato, and then to just as gradually widen the
vibrato until it’s an appoggiatura within a semitone, then a wider
portamento eventually into a wide-ranging glissando. Here we come in about
halfway through this transition and work our way back to the beginning.
I’ve also smoothed over most of her bow changes and introduced a
gradually narrowing filter until she disappears.
i: But you still seem to refer to it as Ms. Jeanrenaud’s performance.
o: I hope that none of the sources completely lose their
original character in the transformation. It’s difficult for me to tell
because I have an intimate relationship to the history of each sound. But
that’s what I’m trying to do — retain some semblance of the
original event. There was something quite hypnotizing and dreamy about Joan
playing that A.
i: But she’s not responding to the rest of the
music we hear.
o: No. She was recorded in isolation. Almost all the
recording I do is of independent elements, which are then flown into a mix.
Any apparent correspondences are fortuitous.
i: You mentioned that there’s a transition to the
nocturnal here. In fact the whole piece becomes quite dreamy from here on
in.
o: And that’s probably the major reason I changed
the ending. The original climactic chord [B.31:11] was like an alarm clock
going off. I’d rather leave the audience to sleep in if that’s
where they’ve gone.
Once we get past the giant hummingbird’s investigation
[A.18:04->18:33] which is a reminiscence of the coneflapper (and then that
filtered solo viola tremolo [A.19:30] which is the last gesture of flight,
now abstract) then, indeed, everything conspires to dreaminess. Layers of
elements are added for their similarity to the existing soundfield. The piano
notes naturally occur farther and farther apart and become progressively more
mellow.
In aparanthesi A I begin to add something to some
of the piano notes, first at note #77 [A.20:11] where I take the note which
has already been tuned down and add its reflection an octave below. Next [at
A.21:41] the octave above and below are added. It doesn’t happen with
any regularity but over the course of five minutes each succeeding chosen
note is reinforced with yet another octave reflection of it, so that [at
A.25:00] there’s an A played in ten octaves. It’s very different
from the eight octave chord at the beginning of the piece because, instead of
being the original eight A’s the piano possesses, it is only one note
in ten octaves, a very different sound.
At this point the piece begins to wind down, or fade. This takes another
five minutes. Or, in the case of B, there’s the penultimate piano note
[at B.28:12] which initiates the whistle breath, and then, following the, in
a sense, highest piano note there’s a reduction of elements
[B.28:35->30:00] leading to a gradual crescendo of a smoothed over
orchestra tuning, which leads up to a reiteration of the initial chord, only
bigger!
Therein lies a problem. I don’t pre-announce that I will attempt to
make this chord big but I think there’s an expectation built in here.
After an audition of aparanthesi B at Naut Humon’s studio in San Francisco, Jon Leidecker said the last chord didn’t seem loud enough,
and I agreed, but any louder and Naut’s system would have distorted.
And in this case there wasn’t the set up of the huge initial chord in
an empty room with the audience outside; the proportions were about the same
as on this CD. Something else is needed here. Anyway I went in a different
direction with aparanthesi A.
i: That last chord lingers for over four minutes.
o: I’ve often ended pieces with unusually long
decays. Die Offnüng [1992] the ballet Bill T Jones and I did in Berlin ends with a diminuendo chord
which lasts six minutes. When I did an orchestral transcription of this
[Classic, 1994], members of the orchestra left the stage one by one
throughout this diminuendo.
Then there’s the challenge of fading out all the way. Most digital
recordings contain something called dither. This is a broadband fuzzy noise
that’s always on at the lower threshold of the CD’s dynamic
range. It helps alleviate low level distortion. In aparanthesi B the decay of the final chord eventually goes below
the threshold of the dither and becomes inaudible and then the dither shuts
off and the CD is over [B.35.33]. You can hear this if you play the very end
of the CD very loud.
Years ago I made a transition between the first track and the second track
on the CD Discosphere [1991] which I found very satisfying. The first piece,
Skindling Shadés slowly fades down, as the fire
depicted dies away; it fades down to digital zero, and as the last ember
fades and the last bit goes goes to zero the first bit of the second piece
Amina begins from digital zero in a slow fade up.
I fussed over the final decay of aparanthesi for
several weeks. While Jean-François Denis wondered if
I was ever going to deliver this music, I kept adding layers of complexity to
what is simply the going away of this drone. There are probably about a dozen
elements that I’m slowly removing and each one has its own independent
fade curve, including the dither. [At A.25:15] you can barely hear the
piano’s sustain sustain pedal lift. Then [from A.26:09->26:23]
there’s the last thing I added which is me whistling for one breath on
the edge of forming an audible note. It’s erratic, it’s the
closest thing to a melody in the whole piece. It’s a bit like a timid
version of Francis Bacon tossing a glob of paint at
a portrait when it’s near finished. It’s a causality casualty
gesture which seals the fate of the piece.
i: I seemed to have missed hearing this.
o: All this is now happening at the threshold of
audibility. Perhaps I’m the only one who will ever hear these things or
notice them. Nonetheless, they’re there. After that last breath
it’s just a multiplicity of congruent fades for about three and a half
minutes [to A.30:00].